This proposal describes the rational design of antitumor agents by exploiting biochemical differences between normal and tumor cells. Certain tumor cells possess high levels of the purine de novo synthetic enzymes IMP dehydrogenase and adenylosuccinate synthetase, as well as a low reduction potential environment. The strategy is to design inhibitors directed towards these enzymes which act only (or preferably) in this low potential environment. A parallel effort will be to design crosslinkers of DNA that do likewise. The design to be employed for both cellular targets is based on a class of naturally occurring quinones known as reductive alkylators (e.g. mitomycin C, the saframycins, and the anthracyclines). These compounds become alkylators when reduced to the hydroquinone form, presumably as a result of quinone methide formation upon elimination of a leaving group. Reductive alkylators based on the imidazo(4,5-g)quinazoline ring system have been shown to mimick murine substrates in enzymatic reactions as well as form an alkylating quinone methide species upon reduction. Thus, a reductive alkylator for xanthine oxidase was designed which acts as a substrate in the oxidized form but inactivates the enzyme upon reduction. The proposed research involves designing nucleotide reductive alkylators of the title en based on the imidazo(4,5 g)quinazoline ring system. Another proposed area of study is the design of mechanistic probes of the title enzymes employing this ring system. This aspect of the project will involve synthesis, enzyme kinetic studies and antitumor screening. Quinone methides based on the benzimidazole and quinazoline ring systems have also been documented and found to be excellent nucleophile traps. It appears that the presence of nitrogen atoms in the quinone methide favor nucleophilic addition over electrophilic addition. Effective alkylators of DNA may be realized by replacing the indole nucleus of the mitosene form of mitomycin C with benzimidazole, quinazoline and other heterocyclic rings. The proposed research involves the synthesis of the mitosene-like systems, DNA crosslinking studies, and antitumor screening. A DNA fragment of defined sequence will be employed to investigate the site of crosslinking.